Encyclopedia of Lunar Science

Living Edition
| Editors: Brian Cudnik

Lunar Interior, Geophysical Models

  • Walter S. KieferEmail author
Living reference work entry
DOI: https://doi.org/10.1007/978-3-319-05546-6_79-1


Our understanding of the Moon’s internal structure is based primarily on the results from experiments deployed by the Apollo 11–17 missions. Most of these experiments were operated from 1969 to 1977, although the laser ranging network continues to be used. The best measurements of internal structure are derived from seismic studies, which provide information about the variation of seismic velocity and mineralogy with depth. Electromagnetic sounding using a combination of surface and orbital magnetometers provides constraints on how electrical conductivity varies with depth in the Moon; in turn, this helps to constrain some aspects of core and mantle composition, particularly metal and water content, along with temperature. Laser ranging measures how the Moon deforms and changes its rotation rate in response to tidal forces; these changes are related to the distribution of mechanically weak layers inside the Moon. Measurements of the Moon’s heat flow determine the rate at which energy flows through the lunar interior, which constrains temperature as a function of depth and thermal evolution over time. Finally, gravity and topography observations from spacecraft in lunar orbit help to extrapolate observations from specific Apollo landing sites to the Moon as a whole.

Crustal Structure and Seismicity

The structure of the Moon’s crust, mantle, and core is constrained primarily by seismic velocities measured using the Apollo Passive Seismometer Experiment, deployed at the Apollo 12, 14, 15, and 16 landing sites. Seismic sources include 1744 meteorite impacts, nine artificial impact events caused by the deliberate crashes of Apollo Saturn rocket upper stages and Lunar Module ascent stages, 28 shallow moonquakes, and 7083 deep moonquakes (Khan et al. 2013). For seismic events that are measured at multiple landing sites, it is possible to use these measurements to determine both the location of the seismic events and how seismic velocity varies with depth below the surface. Figure 1 shows the locations of the best-measured deep and shallow moonquakes, and Fig. 2 shows the seismic velocity structure derived from these observations.
Fig. 1

A cross-sectional view of the Moon’s seismic structure. Shallow moonquakes (black triangles) are from Nakamura et al. (1979) and deep moonquake clusters (red diamonds) are from Nakamura (2005). The solid inner core (black), fluid outer core (red), and partially molten lower mantle (yellow) are from Weber et al. (2011). The blue circles show the locations of the Apollo seismic stations, and the gray lines are depth markers at 500 km and 1000 km below the surface

Fig. 2

Seismic velocities as a function of radius in the Moon. From the surface down to a radius of 480 km (depth 1257 km), the results are based on Gagnepain-Beyneix et al. (2006). The structure in the lowermost mantle (below radius 480 km) and in the core is based on Weber et al. (2011)

The Moon’s crust is an anorthosite-rich layer that formed as a floatation product from the Moon’s magma ocean, covered in some places by a thin veneer (few kilometers thick) of mare basalt. Apollo seismic observations were originally interpreted as indicating an average crustal thickness of 60 km (see entry on “Lunar Magma Ocean Theory, Origins, and Rationale”). Recent reprocessing of the data now favors thinner values, with the P-wave velocity rising from 3.2 km/s at 1 km depth to 7.6 km/s near 30 km depth (the base of the crust; Fig. 2). The seismic measurements determine the crust’s structure only in the vicinity of the Apollo 12, 14, 15, and 16 landing sites, but gravity observations from the Gravity Recovery and Interior Laboratory (GRAIL) mission can now be used to constrain the Moon’s global crustal structure (Wieczorek et al. 2013). The resulting crustal thickness map has an average value of 34–43 km. The crust thins to essentially zero in the centers of the Crisium and Moscoviense impact basins and has a maximum thickness of 80 km in the central part of the Moon’s farside.

The gravity observations also constrain the Moon’s average highland crustal density and porosity, 2,550 kg m−3 and 12%, respectively (Wieczorek et al. 2013). Typical density variations in the highland crust are 100–200 kg m−3 relative to the mean density (Jansen et al. 2017). The porosity is highest, 18–20%, in the ejecta blankets surrounding large impact basins such as Orientale and Moscoviense. The crust’s porosity decreases with depth due to a combination of thermal annealing and overburden pressure, approaching zero at a depth of 20–25 km. This likely explains much of the increase in seismic velocity with depth in the crust (Besserer et al. 2014), although a change from an anorthosite-rich shallow crust to a more mafic composition in the deep crust may also contribute to the seismic velocity change (Khan et al. 2013). The high near-surface porosity is a result of fracturing during the intense impact bombardment experienced by the Moon, particularly during the first 800 million years of its history. Because mare basalts are substantially denser than lunar highland rocks, gravity data can also be used to map subsurface volcanic structures. The magma chamber (now solidified) that fed the Marius Hills volcanic dome field was 200–250 km across and 3–6 km thick (Kiefer 2013). Mapping of buried mare basalts, termed cryptovolcanism, shows that about a third of all lunar mare basalt volcanism was produced prior to 3.8 Ga (Sori et al. 2016). GRAIL data has also provided improved constraints on impact processes, including the thickness of the impact melt sheet and the presence of ring faults at the Orientale basin (Zuber et al. 2016) and the total number and global distribution of large impact basins (Neumann et al. 2015).

Shallow moonquakes are the largest type of internal seismic event detected on the Moon. They occur within about 200 km of the lunar surface, in the crust or upper mantle, although their precise depths are poorly determined by existing data. These are sometimes referred to in the lunar geophysics literature as high frequency teleseisms (HFTs) because these events have an unusually large amount of seismic energy at frequencies greater than 5 Hz. Twenty-eight HFTs were measured by the Apollo seismic network, corresponding to about four per year. The largest measured event had a body wave magnitude of 4.2, but based on the observed size–frequency relationship, larger events may occur occasionally. The triggering mechanism is not known but is not correlated with lunar tides (Nakamura et al. 1979). Thermal moonquakes are small seismic events located at the Moon’s surface that are caused by thermal stresses from the Moon’s diurnal heating and cooling cycle. Individual thermal moonquakes release very little energy, but they may collectively contribute to slope motion and landform degradation (Khan et al. 2013).

Mantle Structure and Seismicity

Seismic velocities show little variation with depth in the lunar mantle (Fig. 2; Gagnepain-Beyneix et al. 2006). This model is parameterized into a small number of layers, reflecting the limited number of constraints on the structure. Because the Apollo seismic network was concentrated in the central part of the Moon’s near side, this seismic velocity model also applies primarily to that portion of the Moon. The observed seismic velocities are consistent with a mantle that is predominantly olivine (~55%) and orthopyroxene (~30%), with the remainder being roughly equal amounts of clinopyroxene and garnet (Khan et al. 2006). The predominance of olivine and orthopyroxene in the mantle is expected from experimental simulations of magma ocean solidification (Elardo et al. 2011). The abundance of garnet depends at least in part on the initial ocean depth, with higher pressures (greater ocean depths) favoring the presence of garnet. The Elardo et al. (2011) experiments at the highest pressure (4 GPa) contain garnet.

The jump in P wave velocity from 7.62 km/s to 8.15 km/s at R = 1000 km is above the uncertainty of 0.22 km/s in the velocities and might be due to a change in chemical composition at this depth. A chemically layered mantle is a possible outcome of solidification of the Moon’s magma ocean. Because of the limited number of layers in the parameterization of the seismic velocity structure, the precise depth of the compositional change is not well determined, and some other studies have suggested that it occurs at shallower depths. On the other hand, the S wave velocity change from 4.4 km/s to 4.5 km/s at the same depth does not represent a clear change given the velocity uncertainty of 0.10 km/s. Moreover, an alternative parameterization of the velocity structure can explain the seismic data without a discrete jump in P wave velocity (Garcia et al. 2011), so the possibility of chemical stratification in the lunar mantle remains unresolved. An improved knowledge of the seismic velocity structure in the middle and lower mantle is an important objective for a future lunar geophysics mission and would contribute significantly to understanding how the lunar magma ocean differentiated. The upper part of the lunar mantle has very low seismic attenuation (Garcia et al. 2011), with a P-wave seismic attenuation quality factor exceeding 1,000 in the upper 700 km. This requires a cold, dry upper mantle, with a temperature far below that of the solidus; up to 10–100 ppm water might be present, depending on the temperature (Karato 2013).

Deep moonquakes (DMQs), with body wave magnitudes of one to three, typically occur at depths of 700–1100 km below the surface (Fig. 1) and are located in more than 300 clusters of events (Nakamura 2005). Different seismic events in a given DMQ cluster have very similar seismic waveforms, indicating that each event has the same focal mechanism, with hypocenters located in a zone only 1–2 km across. In many of these clusters, seismic events occur periodically in time, with a repeat period ~27 days that is indicative of forcing by tidal stresses, although for some DMQ clusters other, poorly understood forcing mechanisms may also contribute (Bulow et al. 2007; Weber et al. 2009). A puzzling aspect of DMQs is their association with stress variations of only about 0.01 MPa, which is extremely low considering that the hydrostatic pressure at 700 km depth is 3 GPa. Of the well located DMQ clusters, 98 occur on the Moon’s near side, spatially clustered in the vicinity of lunar mare basalts, and only two to eight occur on the Moon’s farside (Nakamura 2005; see Fig. 1). It is possible that the spatial distribution and stress levels are related, because deep mantle water associated with the source region of the Moon’s mare basalts might cause localized weakening of the mantle, which could in turn facilitate DMQ activity (Qin et al. 2012). On the other hand, DMQs might occur throughout the lunar far side without ever being seen on the Apollo seismic network because the ray paths from such seismic events would traverse the high seismic attenuation zone in the Moon’s deep mantle.

The Core and the Core-Mantle Boundary

Weber et al. (2011) and Garcia et al. (2011) reported evidence for P- and S-waves from DMQs that were reflected off of the Moon’s core prior to being detected at the lunar surface. Reflected seismic waves of this type can occur where there are large changes in seismic velocity with depth, so these investigators interpreted their results in terms of reflections off of the Moon’s core. Weber et al. (2011) identified three such reflective layers in their results (Fig. 2). They interpreted a layer at radius R = 330 ± 20 km as being the top of the core and a layer at R = 240 ± 10 km as indicating the Moon’s inner core. In this interpretation, the outer core is liquid and the inner core is solid. They also identified a reflecting layer at R = 480 ± 15 km and interpreted the region between R = 330 km and 480 km as a partially molten silicate layer at the base of the mantle. Based on the relatively small decrease in seismic velocities in the R = 330–480 km zone, there is thought to be between 5% and 30% partial melt in this layer. Independent analysis of a similar data set by Garcia et al. (2011) implies a slightly larger core radius, R = 380 ± 40 km, but does not require either an inner core or a partially molten mantle layer.

Independent constraints on core structure come from electromagnetic sounding, laser ranging, and the lunar gravity field. Electromagnetic sounding with orbital magnetometers indicates the presence of a region of very high electrical conductivity, interpreted as a metal-rich core. Hood et al. (1999) estimated a core radius of 340 ± 90 km using Lunar Prospector spacecraft data, and Shimizu et al. (2013) estimated a core radius of 290 km using Kaguya spacecraft data, with lower and upper limits of 170 km and 400 km. Williams et al. (2014) and Matsuyama et al. (2016) combined laser ranging measurements with GRAIL observations of the Moon’s gravity field to measure the tidal Love number, the monthly tidal dissipation, and the moment of inertia. The tidal Love number, k2, measures how the Moon deforms in response to tidal forces, which is related to the distribution of mechanically weak layers, such as regions that are molten or partially molten. The monthly tidal dissipation measures the rate at which the Moon’s rotation is slowing down; tidal dissipation occurs most effectively due to friction at the boundary between liquid and solid regions. The moment of inertia measures the extent to which mass is concentrated near the center of the Moon. The combined results require the presence of a dense core with a radius between 200 km and 400 km (Williams et al. 2014; Matsuyama et al. 2016). The observed tidal dissipation requires that at there is at least a thin liquid outer core layer about 100 km thick on top of the solid inner core. The value of k2 observed by GRAIL permits the existence of a mantle layer with low rigidity and low seismic velocity (Matsuyama et al. (2016) refer to this as the “transition layer”), but the existence of this layer is not required by current k2 data. If this transition layer does exist in the lowermost mantle, it extends from the top of the outer core to R ~ 500 km, similar to the low seismic velocity, partially molten silicate layer inferred by Weber et al. (2011) on the basis of seismic travel times.

These results require measuring very weak signals and the overall concordance of results using four different data types is impressive. These results demonstrate that the Moon has a small, iron-rich core, corresponding to 1.0–2.0% of the Moon’s mass. This overlaps with geochemical estimates from siderophile element abundances that favor a lunar core mass of 0.7–2.5% of the Moon’s mass (Righter 2002; Rai and van Westrenen 2014). For comparison, the Earth’s core is 31% of Earth’s mass. The small size of the Moon’s core is an important constraint on the processes that formed the Moon, such as the proposed giant impact model. At least part of the core is currently fluid; the presence of a distinct solid inner core and of a partially molten lowermost mantle is suggested by these observations but is not yet definitively measured. If a future geophysical network can confirm the detailed structure of the inner and outer core and of a partially molten lowermost mantle, such observations will provide important constraints on the Moon’s present-day thermal structure. The structure, composition, and thermal state of the core are also important for understanding the Moon’s magnetic dynamo history (see entry on “Lunar Core Dynamo”).

Heat Flow and Thermal Structure

The Apollo Heat Flow Experiment measured the thermal gradient and the thermal conductivity in the upper 2–3 m of the Moon. The product of these two quantities determines the flux of energy out of the Moon and can be used to constrain the deeper thermal structure of the Moon. Unfortunately, the measured values at the Apollo 15 (21 ± 3 mW m−2) and 17 (16 ± 2 mW m−2) landing sites are unlikely to be representative of the Moon as a whole. Both landing sites are on the edges of major impact basins, and the large change in crustal thickness in these locations likely leads to some degree of lateral focusing of thermal energy, making the heat flow measured at these locations higher than elsewhere on the Moon. Moreover, the Apollo 15 measurements and to a lesser extent the Apollo 17 measurements are affected by the presence of the high radioactivity Procellarum KREEP Terrane (PKT), which also raises the heat flow in these locations relative to the Moon as a whole. Siegler and Smrekar (2014) estimated that the Moon’s globally averaged mantle heat flux is 7–13 mW m−2, although this needs to be tested by direct measurements in regions that are far from both the PKT and the edges of impact basins.

Numerical models that include the effects of heat transport by both thermal conduction and mantle convection, as well as partitioning of radioactivity between the PKT and the rest of the Moon, can be used to estimate both the temperature as a function of depth in the present-day Moon as well as how the temperature has cooled over time (Laneuville et al. 2013; see entry on “Internal Structure/Mantle Motions of the Moon”). The lunar magma ocean is expected to solidify quickly, with 80% solidification in just 1000 years and complete solidification in less than 10 million years (Elkins-Tanton et al. 2011). For this reason, the long-term thermal evolution models of Laneuville et al. (2013) assumed that the upper portion of the lunar mantle was initially at its solidus temperature. These models predict that the deep interior is hot enough that at least part of the core will be liquid if it contains a few percent of sulfur or carbon, which is chemically likely. However, such models predict that the deep mantle is too cold to be partially molten at present if the mantle composition is homogeneous. In order to maintain melt at the base of the mantle over the age of the Solar System, an additional source of energy in the deep interior is necessary. It is likely that the Moon passed through a magma ocean phase, and during the solidification of the ocean a layer of dense, ilmenite-rich material could have formed at the base of the mantle. Such a layer would also be enriched in rare earth elements, including the radioactive elements uranium and thorium, which could provide an extra heat source to maintain the partially molten deep mantle layer. In addition, tidal heating would be concentrated in the low viscosity, partially molten silicates, which would provide another possible heat source for maintaining the partially molten layer.

Future Directions

Future advances in our understanding of the Moon’s interior structure are likely to require deployment of a new generation of geophysical instruments on the Moon. NASA is studying a Lunar Geophysical Network for possible launch in the 2020s (National Research Council 2011). This network would include landers at four or more locations on the Moon and would expand the geographic range of the Apollo experiment array by locating stations near the Moon’s limb as well as on the far side. Each lander would carry a broadband seismometer, a heat flow probe, and an electromagnetic sounding experiment. Landers located on the Moon’s near side would also carry a laser ranging retroreflector. Such a network would improve our knowledge of seismic velocities in the deep Moon (lower mantle and core) and would definitively determine whether deep moon quakes are localized on the lunar near side or are globally distributed. Moreover, such a network would improve our knowledge of the Moon’s thermal evolution by measuring the average heat flow and by determining the degree of variation of heat flow inside and outside the Procellarum KREEP Terrane.




  1. Besserer J, Nimmo F, Wieczorek MA et al (2014) GRAIL gravity constraints on the vertical and lateral density structure of the lunar crust. Geophys Res Lett 41:5771–5777ADSCrossRefGoogle Scholar
  2. Bulow RC, Johnson CL, Bills BG, Shearer PM (2007) Temporal and spatial properties of some deep moonquake clusters. J Geophys Res 112:E09003.  https://doi.org/10.1029/2006JE002847 ADSCrossRefGoogle Scholar
  3. Elardo SM, Draper DS, Shearer CK (2011) Lunar magma ocean crystallization revisited: bulk composition, early cumulate mineralogy, and the source regions of the highlands Mg-suite. Geochim Cosmochim Acta 75:3024–3045ADSCrossRefGoogle Scholar
  4. Elkins-Tanton LT, Burgess S, Yin Q-Z (2011) The lunar magma ocean: reconciling the solidification process with lunar petrology and geochemistry. Earth Planet Sci Lett 304:326–336ADSCrossRefGoogle Scholar
  5. Gagnepain-Beyneix J, Lognonné P, Chenet H et al (2006) A seismic model of the lunar mantle and constraints on temperature and mineralogy. Phys Earth Planet Int 159:140–166ADSCrossRefGoogle Scholar
  6. Garcia RF, Gagnepain-Beyneix J, Chevrot S, Lognonné P (2011) Very preliminary reference Moon model. Phys Earth Planet Int 188:96–113ADSCrossRefGoogle Scholar
  7. Hood LL, Mitchell DL, Lin RP et al (1999) Initial measurements of the lunar induced magnetic dipole moment using Lunar Prospector Magnetometer data. Geophys Res Lett 26:2327–2330ADSCrossRefGoogle Scholar
  8. Jansen JC, Andrews-Hanna JC, Li Y et al (2017) Small-scale density variations in the lunar crust revealed by GRAIL. Icarus 291:107–123ADSCrossRefGoogle Scholar
  9. Karato S (2013) Geophysical constraints on the water content of the lunar mantle and its implications for the origin of the Moon. Earth Planet Sci Lett 384:144–153ADSCrossRefGoogle Scholar
  10. Khan A, Maclennan J, Taylor SR, Connolly JAD (2006) Are the Earth and Moon compositionally alike? Inferences on lunar composition and implications for lunar origin and evolution from geophysical modeling. J Geophys Res 111:E05005.  https://doi.org/10.1029/2005JE002608 ADSGoogle Scholar
  11. Khan A, Pommier A, Neumann GA, Mosegaard K (2013) The lunar moho and the internal structure of the Moon: a geophysical perspective. Tectonophys 609:331–352CrossRefGoogle Scholar
  12. Kiefer WS (2013) Gravity constraints on the subsurface structure of the Marius Hills: the magmagic plumbing of the largest lunar volcanic dome complex. J Geophys Res: Planets 118:733–745ADSCrossRefGoogle Scholar
  13. Laneuville M, Wieczorek MA, Breuer D, Tosi N (2013) Asymmetric thermal evolution of the Moon. J Geophys Res: Planets 118:1435–1452ADSCrossRefGoogle Scholar
  14. Matsuyama I, Nimmo F, Keane JT et al (2016) GRAIL, LLR, and LOLA constraints on the interior structure of the Moon. Geophys Res Lett 43:8365–8375ADSCrossRefGoogle Scholar
  15. Nakamura Y (2005) Farside deep moonquakes and deep interior of the Moon. J Geophys Res 110:E01001.  https://doi.org/10.1029/2004JE002332
  16. Nakamura Y, Latham GV, Dorman HJ et al (1979) Shallow moonquakes: depth, distribution and implications as to the present state of the lunar interior. Proc Lunar Planet Sci Conf 10:2299–2309ADSGoogle Scholar
  17. National Research Council Committee on the Planetary Science Decadal Survey (2011) Vision and voyages for planetary science in the decade 2013–2022. National Research Council, Washington, DC. 382 pagesGoogle Scholar
  18. Neumann GA, Zuber MT, Wieczorek MA et al (2015) Lunar impact basins revealed by Gravity Recovery and Interior Laboratory measurements. Sci Adv 1:e1500852.  https://doi.org/10.1126/sciadv.1500852 ADSCrossRefGoogle Scholar
  19. Qin C, Muirhead AC, Zhong S (2012) Correlation of deep moonquakes and mare basalts: implications for lunar mantle structure and evolution. Icarus 220:100–105ADSCrossRefGoogle Scholar
  20. Rai N, van Westrenen W (2014) Lunar core formation: new constraints from metal–silicate partitioning of siderophile elements. Earth Planet Sci Lett 388:343–352ADSCrossRefGoogle Scholar
  21. Righter K (2002) Does the Moon have a metallic core? Icarus 158:1–13ADSCrossRefGoogle Scholar
  22. Shearer CK, Hess PC, Wieczorek MA et al (2006) Thermal and magmatic evolution of the Moon. In: New views of moon, Reviews in mineralogy and geochemistry, Mineralogical Society of America Chantilly, Virginia USA, vol 60, pp 365–518Google Scholar
  23. Shimizu H, Matsushima M, Takahashi F et al (2013) Constraint on the lunar core size from electromagnetic sounding based on magnetic field observations by an orbiting satellite. Icarus 222:32–43ADSCrossRefGoogle Scholar
  24. Siegler MA, Smrekar SE (2014) Lunar heat flow: regional prospective of the Apollo landing sites. J Geophys Res: Planets 119:47–63ADSCrossRefGoogle Scholar
  25. Sori MM, Zuber MT, Head JW, Kiefer WS (2016) Gravitational search for cryptovolcanism on the Moon: evidence for large volumes of early igneous activity. Icarus 273:284–295ADSCrossRefGoogle Scholar
  26. Weber RC, Bills BG, Johnson CL (2009) Constraints on deep moonquake focal mechanisms through analyses of tidal stress. J Geophys Res 114:E05001.  https://doi.org/10.1029/2008JE003286 ADSCrossRefGoogle Scholar
  27. Weber RC, Lin P, Garnero EJ et al (2011) Seismic detection of the lunar core. Science 331:309–312ADSCrossRefGoogle Scholar
  28. Wieczorek MA, Jolliff BL, Khan A et al (2006) The constitution and structure of the lunar interior. In: New views of moon, Reviews in mineralogy and geochemistry, Mineralogical Society of America Chantilly, Virginia USA, vol 60, pp 221–364Google Scholar
  29. Wieczorek MA, Neumann GA, Nimmo F et al (2013) The crust of the Moon as seen by GRAIL. Science 339:671–675ADSCrossRefGoogle Scholar
  30. Williams JG, Konopliv AS, Boggs DH et al (2014) Lunar interior properties from the GRAIL mission. J Geophys Res: Planets 119:1546–1578ADSCrossRefGoogle Scholar
  31. Zuber MT, Smith DE, Neumann GA et al (2016) Gravity field of the Orientale basin from the Gravity Recovery and Interior Laboratory mission. Science 354:438–441ADSCrossRefGoogle Scholar

Copyright information

© Springer International Publishing AG 2018

Authors and Affiliations

  1. 1.Lunar and Planetary Institute/USRAHoustonUSA

Section editors and affiliations

  • Edgar Sikko Steenstra
    • 1
  1. 1.Faculty of Earth and Life SciencesVrije UniversiteitAmsterdamThe Netherlands